Wages growth can typically be linked to the labour market measures like employment growth six months earlier.

In the final quarter of 2007, for example, annual employment growth was running strongly at 3.2 per cent, before being ground down to just 0.2 per cent by the second quarter of 2009.

Growth in the wage price index produced, like the labour force data, by the Australian Bureau of Statistics, peaked at 4.3 per cent in the second quarter of 2008, half a year after jobs growth peaked, and bottomed out at 2.9 per cent, two quarters after jobs growth.

The link between jobs and wages is never perfectly regular, but it appears to have gone completely haywire over the past year or so.

The recent low point for annual jobs growth – using quarterly average data – was in the final quarter of 2013.

Since then annual jobs growth has picked up from 0.5 per cent to 2.1 per cent.

But wages growth has continued to slow, from 2.5 per cent two quarters after jobs growth bottomed, when wages growth ought to have hit its low point, to just 2.3 per cent in the second quarter of 2015, according to figures released by the bureau on Wednesday.

The wage price index only takes us back to 1997, but other longer-running data sets, like average weekly earnings and the average employee compensation measure from the national accounts, suggest wages growth is at its slowest for at least half a century, with the possible exception of a few quarters during the early-1990s recession.

Even though jobs growth really ought to be picking up.

Three possible explanations spring to mind.

One is that weakness in specific sectors, notably mining and the mining-related construction and services, is weighing the average down as the mining investment boom goes bust.

Another is that, although jobs growth is back up to about normal, there is a residual drag on wages from the prior three years of below-par jobs growth.

A third is that jobs growth is simply being overestimated.

It wouldn’t be the first time.

A possible culprit is slower immigration.

But the bureau’s assumption for annual growth in the estimated working-age population, which typically moves in lockstep with migration and long-term visitor flows – and which forms the basis for employment estimates, has been virtually unchanged for the past four years.

If population growth is being overestimated, as seems likely, then jobs growth probably is, too.

But just which of these three possible factors, or what combination of them, might resolve the wages/jobs conundrum remains to be established.